Hi there,
Thanks a lot for releasing this valuable dataset. By the way, I have a few questions:
1) In the paper events are annotated by triggers, but in this dataset there are no triggers. Are this event extraction task actually a trigger-free task?
2) In the paper you mentioned the pos/neg sentence to represent event mentions. But the dataset doesn't contain pos/neg flag. Should I segment each document into natural sentences by myself for recognizing them?
I would appreciate it if you could answer my questions. Thx!
Hi there, Thanks a lot for releasing this valuable dataset. By the way, I have a few questions: 1) In the paper events are annotated by triggers, but in this dataset there are no triggers. Are this event extraction task actually a trigger-free task? 2) In the paper you mentioned the pos/neg sentence to represent event mentions. But the dataset doesn't contain pos/neg flag. Should I segment each document into natural sentences by myself for recognizing them? I would appreciate it if you could answer my questions. Thx!